DeepSeek participated in the strategic placement of Unitree Technology with approximately RMB 141 million. Cui Tianyi, head of Harness, publicly questioned the description of DeepSeek V4’s programming capabilities in the related IPO materials. The materials claimed that DeepSeek V4’s programming ability was only 0.3% weaker than Claude’s flagship model. Cui pointed out that, according to the V4 paper, only the SWE Verified benchmark showed performance close to Opus 4.6 and could not be used to generalize the overall programming capability gap, calling this wording misleading.
DeepSeek's IPO materials face criticism for exaggerated claims about programming capabilities
AiCoinShare
Summary
On-chain news reports that DeepSeek’s IPO materials have faced criticism for exaggerating the programming performance of DeepSeek V4. Harness leader Cui Tianyi noted that the materials claim V4 is only 0.3% behind Claude’s top model, but the V4 paper only shows a SWE Verified score near Opus 4.6, which does not fully reflect its programming capabilities. Crypto news outlets are highlighting the debate as investors closely examine technical claims in the latest token project developments.
Source:Show original
Disclaimer: The information on this page may have been obtained from third parties and does not necessarily reflect the views or opinions of KuCoin. This content is provided for general informational purposes only, without any representation or warranty of any kind, nor shall it be construed as financial or investment advice. KuCoin shall not be liable for any errors or omissions, or for any outcomes resulting from the use of this information.
Investments in digital assets can be risky. Please carefully evaluate the risks of a product and your risk tolerance based on your own financial circumstances. For more information, please refer to our Terms of Use and Risk Disclosure.